AI Visibility Reports You Can Defend

You have seen the report. A big number, a few colored bars, a line about “AI visibility” that nobody can point to. It looks authoritative and it is quietly indefensible — the first time a client asks “how did you get this number?”, the whole thing wobbles. That is the feeling this page is about: reports that read like they were made up, because the evidence behind them was never in the report. A defensible visibility report is not one with a higher score. It is one where every finding can be re-checked, the method is written down, verified signals are separated from heuristics, and the limits are stated out loud.

Key takeaways

A single proprietary “visibility score” with no traceable basis is the reason these reports feel made up.
Google says no third-party tool has access to its internal ranking or AI systems — treat any tool that implies it does as a red flag.
Defensible = every finding tied to evidence you can re-check, a published methodology, and verified vs heuristic labels kept separate.
State what is not measured (no citation tracking, no share of voice, no observed crawler hits) before a client finds the gap for you.
No report can guarantee rankings, traffic, or citations — Google itself says indexing and serving are not guaranteed.

Why the reports feel made up

VERIFIED FACT — Google's own guidance on third-party tools

The uncomfortable truth is that the headline metric in most AI-visibility dashboards cannot be independently verified, because it is not derived from anything public. Google Search Central is explicit about this in its guide to optimizing for generative AI search: “Be wary of third-party tools that promise ranking success or claim to use ‘internal’ Google metrics. No third-party tool has access to our internal ranking or AI systems.” You can read the source directly at Google Search Central. When a report presents a precise-looking “AI visibility index” as if it were measured, and there is no evidence you can re-check, the report is not wrong so much as unverifiable — and unverifiable is what “made up” feels like.

What a defensible report contains

EVIDENCE-BACKED RECOMMENDATION — the four things to demand from any visibility report you put your name on

Evidence per finding. Not a score — the actual thing found: the robots.txt directive, the meta robots value, the JSON-LD field, the HTTP status, the evaluated URL and user-agent. If a person cannot re-check it, it does not belong in a client report.
A published methodology. A written page that says what is measured, how evidence is evaluated, where human judgment is applied, and what is out of scope. Ours is public at /methodology/visibility.
Verified vs heuristic labels. Deterministic findings and pattern-based signals are kept separate and labeled, never averaged into one number presented as fact.
Stated limits.A clear “what we do not measure” section, so the report never overclaims in a room where you have to defend it.

Verified vs heuristic, in practice

This is the distinction that saves you in a client meeting. Here is how the same question looks at each evidence level.

Verified finding

Deterministic and re-checkable. The page either exposes a valid FAQPage JSON-LD block or it does not. robots.txt either permits the crawler path or it does not. The title tag is either within the documented length range or it is not. Anyone can pull the same file and confirm the result.

Heuristic finding

A pattern associated with readiness. Answer-block structure, question-formatted headings, fact density. These are directional signals — the kind of thing that tends to help — not proof that any specific engine will lift or cite the page. AEO Readiness is labeled Beta for exactly this reason.

Blending the two into a single “visibility score” is how a report starts to feel invented: the verified part gets diluted by the heuristic part, and the number stops meaning anything you can stand behind. Optimus Pass keeps them labeled and separate.

What we deliberately do not claim

VERIFIED FACT — scope limits, stated up front

No continuous, automated AI-prompt monitoring on your behalf.
No citation tracking over time, and no “share of voice” percentage.
No claim that a crawler actually visited — crawler-access findings are configuration evidence, not observed activity.
No whole-site crawl coverage claim; a scan evaluates the URLs it retrieved.
No guarantee of ranking, indexing, traffic, AI Overview inclusion, or citations.

Google backs the last point directly: even a page that meets every requirement and best practice is not promised service — “Indexing and serving isn't guaranteed.” (AI features and your website). A report that hides this is the one that eventually embarrasses you in front of a client.

The human layer: Executive Client Visibility Audit

The free scan shows what the engine measures. When a client decision is on the line, the audit adds a person: the full visibility engine runs across one client domain, a founder reviews and interprets the findings, and you receive a polished, client-ready executive report, a prioritized action plan, and a 30-minute findings call. Founding rate: $750 for one client domain.

Run a free visibility scan See the Executive Audit — $750

Start with the scan to see the evidence format for yourself. The audit is for when you need a human read you can put in front of a client.

What Optimus Pass can and cannot do

Optimus Pass evaluates the public, verifiable conditions that affect whether search engines and AI assistants can find, read, and surface a site — and, for a defined prompt set, records what a human observed at a point in time. It does not access any search engine's internal systems, does not track citations over time, and cannot guarantee ranking, traffic, inclusion, or citations. No tool honestly can. Use it to prioritize fixes with evidence, not as a promise of outcomes.

Frequently asked questions

Why do AI visibility reports feel made up?+

Because most of them present a single proprietary score — an 'AI visibility index' or 'share of voice' — with no traceable basis behind it. You cannot see which page was checked, what was found, or how the number was computed. Google Search Central states plainly that no third-party tool has access to its internal ranking or AI systems, and warns against tools that claim to. A report that cannot show its evidence is a number, not a finding.

What makes a visibility report defensible in front of a client?+

Every material finding is tied to evidence a person can re-check: the robots.txt directive, the meta robots tag, the JSON-LD field, the HTTP status, the evaluated URL and user-agent. The methodology is published. Findings are labeled by evidence level — verified vs heuristic — instead of blended into one fake-precise score. And the report states what it does not measure, so you are never surprised in a client meeting.

What is the difference between a verified finding and a heuristic one?+

A verified finding is deterministic and re-checkable: the page either has a valid FAQPage block or it does not; robots.txt either permits the crawler path or it does not. A heuristic finding is a pattern associated with readiness — for example, answer-block structure in AEO Readiness (Beta). Heuristics are directional signals, not proof of an outcome. Optimus Pass keeps the two labeled separately and never merges them into a single 'visibility score' presented as fact.

Does Optimus Pass track AI citations or share of voice?+

No. It does not run continuous automated prompt monitoring, does not track where or whether AI engines cite you over time, and does not compute a share-of-voice percentage. Where AI-answer evidence appears in an Optimus report, it is a point-in-time observation recorded by a human analyst for a defined set of prompts — labeled as a snapshot, not a trend. This is a deliberate limit, not a gap to hide.

Can a visibility report guarantee rankings, traffic, or citations?+

No report can, and anyone who promises that is selling you something. Google states that meeting all requirements and best practices still does not mean a page will be crawled, indexed, or served — indexing and serving are not guaranteed. Optimus Pass reports conditions you can verify and improve. It does not promise outcomes.

What is the Executive Client Visibility Audit?+

A human service for agencies. The same engine runs a full visibility audit across one client domain, then a founder reviews and interprets the findings and delivers a polished, client-ready executive report, a prioritized action plan, and a 30-minute findings call. The founding rate is $750 for one client domain.

Read exactly how each check is evaluated, and what is out of scope:

Visibility Intelligence Methodology →

Your AI visibility report shouldn't feel made up